Development of An Asynchronous Solar-Powered Cooker

1
DEVELOPMENT OF AN ASYNCHRONOUS
SOLAR-POWERED COOKER
by
P. Femi Akinwale
Submitted to the Department of Mechanical Engineering
in Partial Fulfillment of the Requirements
for the Degree of Master of Science in Mechanical Engineering
at the
Massachusetts Institute of Technology
June 2006
2006 Pamela O. Akinwale. All rights reserved
The author hereby grants to MIT permission to reproduce and to distribute publicly paper and
electronic copies of this thesis document in whole or in part in any medium now known or
hereafter created
Signature of Author
Department of Mechanical Engineering
June 5, 2006
Certified by..
David Gordon Wilson
Professor Emeritus of Mechanical Engineering
Thesis Supervisor
Accepted by.
Lallit Anand
Professor of Mechanical Engineering
Chairman, Department Committee on Graduate Students

2
DEVELOPMENT OF AN ASYNCHRONOUS
SOLAR-POWERED COOKER
by
P. Femi Akinwale
Submitted to the Department of Mechanical Engineering
on 5 June 2006
in partial fulfillment of the requirements for the
Degree of Master of Science in Mechanical Engineering
ABSTRACT
One reason that solar cookers have not gained widespread acceptance is because their use has
proved inconvenient and impractical. Users are restricted to cooking when, and where, the sun is
shining. Furthermore, the cooking temperature can not readily be raised or lowered as desired.
In contrast, the Wilson solar cooker is designed to permit use under conditions characterized by
low or no insolation. Furthermore, the design would facilitate users adjusting temperatures. These
temperatures would reach levels as high as 258
0
C. In order to validate the concept, construction of
one prototype was initiated.
Lithium nitrate, the heat-storage material, was shown to meet the stated requirements of storing
heat at a constant temperature of 258
0
C for up to six hours. Furthermore, this heat-storage material
stored heat at temperatures above the boiling point of water, for up to 25 hours. Thus, it is expected
that a meal for six people can be prepared up to six hours after charging of the thermal battery.

Thesis Supervisor: Professor David Gordon Wilson
Title: Professor of Mechanical Engineering

3
TABLE OF CONTENTS
ABSTRACT 2
TABLE OF CONTENTS 3
LIST OF TABLES 6
LIST OF FIGURES 7
LIST OF EQUATIONS 9
NOMENCLATURE 10
A) Symbols 10
B) Subscripts 10
ACKNOWLEDGMENTS 11
GLOSSARY 12
CHAPTER 1) INTRODUCTION 14
A) Mechanism of operation 14
i) Collection 14
ii) Absorption 15
iii) Heat transfer and storage 16
B) Goals 16
i) Deliverables 16
ii) Planned features 17
iii) Deferred features 17
C) Key Topics 17
i) Heat-absorption rate during charging 17
ii) Heat-loss rate during storage 17
iii) Heat-transfer rate during cooking 17
iv) Useful heat 17
v) Maximum systemtemperature 17
CHAPTER 2) BACKGROUND 19
A) History 19
i) Temporal restrictions 20
ii) Spatial restrictions 21
iii) Temperature restrictions 22
TABLE OF CONTENTS

4
B) Components 22
i) Collector 23
ii) Thermal battery 23
iii) Stovetop 24
C) Analysis 24
i) Collection. 24
ii) Thermal charging 25
iii) Thermal dissipation 26
iv) Module geometry design 27
v) Spring design 27
D) Design 27
CHAPTER 3) METHODS 29
A) Material selection 29
i) Heat-storage-material selection 29
ii) Module-material selection 32
iii) Compatibility research 33
iv) Compatibility tests 34
B) Module-shape design 35
i) Module-shape criteria. 35
ii) Module-shape selection 36
C) Module construction 39
i) Material procurement & fabrication 39
ii) Assembly 40
iii) Joining and sealing 41
D) Assembled module tests 42
i) Oven 42
ii) Insulated storage container 43
iii) Temperature measurement 44
E) Module re-design. 45
i) Design cycle two 45
ii) Redesigned module test 47
CHAPTER 4) RESULTS AND DISCUSSION 48
A) Health risks 48
B) Chemical and physical properties 48
i) Test 1: stainless steel cup and mild steel cap 48
ii) Test 2: mild steel flat pouch 49
C) Thermal property observations 49
i) High surface area to volume ratio 49
ii) Low conductivity of lithiumnitrate 49
iii) High latent-heat capacity 50
D) Thermal energy storage results 50
i) Test 3: single steel module 50
ii) Test 4: two aluminum modules 53
TABLE OF CONTENTS

5
iii) Test 5: three aluminum modules 54
E) Key system parameters 55
i) Heat-absorption rate during charging 55
ii) Heat-loss rate during storage 56
iii) Heat-transfer rate during cooking 58
iv) Useful heat 58
v) Maximum systemtemperature 58
CHAPTER 5) CONCLUSION AND RECOMMENDATIONS 59
A) Design 59
i) Materials selection 59
ii) Module construction 59
iii) Module-shape design 60
B) Performance 60
C) Deferred features 61
i) Salt-state indicator 61
ii) Operator control of stovetop heat-transfer rate during cooking. 61
iii) Automatic tracking of the suns apparent movement. 61
iv) Radiant energy collection 61
v) Thermal energy absorption 61
vi) Cooking 61
REFERENCES 63

6
LIST OF TABLES
Number
Table 1: Absorptance ratio and emissivity for selective coatings......................................... 15
Table 2: Module charging time at 50% collection efficiency, with tracking....................... 25
Table 3: Spring geometry....................................................................................................... 27
Table 4: Required volume and mass for selected sensible heat media
5
............................... 31
Table 5: Thermal properties of nitrate phase-change materials............................................ 31
Table 6: Thermal properties of lithium nitrate...................................................................... 31
Table 7: Physical properties of solid lithium nitrate............................................................. 32
Table 8: Physical properties of liquid lithium nitrate............................................................ 32
Table 9: Properties of module-material candidates............................................................... 32
Table 10: Lithium nitrate thermal expansion constants........................................................ 37
Table 11: Module geometry design values (at room temperature)....................................... 38
Table 12: Standard AWS brazing-alloy usage temperatures
9
............................................... 42
Table 13: Heat-loss rate as a function of heat-storage phase................................................ 56
Table 14: Predicted heat-loss rate as a function of temperature difference.......................... 56
Table 15: Heat-loss rate for varying numbers of modules.................................................... 57
Table 16: Predicted heat loss as a function of insulation thickness.
5
................................... 57
Table 17: Heat-loss rate for different insulating materials.................................................... 58
Table 18: Thermal conductivity of common materials......................................................... 61

7
LIST OF FIGURES
Number Page
Figure 1: Tracking box cooker............................................................................................... 19
Figure 2: Fresnel solar cooker................................................................................................ 19
Figure 3: VITA solar cooker.................................................................................................. 20
Figure 4: Parabolic solar cooker............................................................................................ 20
Figure 5: Clear Dome solar cooker........................................................................................ 21
Figure 6: Stored-heat concentrating reflector in Salta, Argentina........................................ 22
Figure 7: Early Wilson solar cooker design.......................................................................... 23
Figure 8: Thermal conduction model .................................................................................... 26
Figure 9: Pipe insulation model ............................................................................................. 26
Figure 10: Recommended Wilson Solar Cooker: Version 1................................................ 28
Figure 11: Recommended Wilson Solar Cooker: Version 2................................................ 28
Figure 12: Proposed Wilson solar cooker design.................................................................. 29
Figure 13: Relative machinability of steels........................................................................... 33
Figure 14: Flat steel pouch test set-up................................................................................... 35
Figure 15: Early module design............................................................................................. 36
Figure 16: Revised module design........................................................................................ 37
Figure 17: Three-dimensional view of the design selected for the module top and bottom38
Figure 18: Module-top design (stainless steel material, units in inches).............................. 39
Figure 19: Manufactured module top.................................................................................... 40
Figure 20: Module bottom..................................................................................................... 40
Figure 21: Insulated storage container................................................................................... 43
Figure 22: Top and side views of the insulated storage container........................................ 43
Figure 23: Metal drum used in construction of the insulated storage container: ................. 44
Figure 24: MDC Copper gasket............................................................................................. 46
LIST OF FIGURES

8
Figure 25: Temperature versus time in a stainless steel module.......................................... 51
Figure 26: Theoretical predictions for temperature versus time in a pure substance cooled at
constant pressure........................................................................................................ 51
Figure 27: Temperature versus time of lithium nitrate in a stainless steel module (after
solidification)............................................................................................................. 52
Figure 28: Temperature difference across a single steel module in sawdust insulation...... 53
Figure 29: Temperature versus time for two modules in a pre-heated storage container.... 54
Figure 30: Temperature versus time for three modules in a pre-heated storage container.. 55

9
LIST OF EQUATIONS
Equation 1: Module charging time........................................................................................ 25
Equation 2: Rate of thermal energy transfer.......................................................................... 26
Equation 3: Time required for thermal energy dissipation................................................... 26
Equation 4: Heat loss through a cylindrical surface.............................................................. 27
Equation 5: Total rate of heat loss through pipe insulation.................................................. 27
Equation 6: Module energy storage....................................................................................... 27
Equation 7: Module temperature difference.......................................................................... 27
Equation 8: Lithium nitrate thermal expansion..................................................................... 37

10
NOMENCLATURE
A) Symbols
A =area m
2
c =specific-heat capacity at constant pressure J /(kg.K)
=difference
E =thermal energy J
h
f
=latent-heat capacity J /kg
I =solar irradiance W/m
2
m =mass of material kg
k =thermal conductivity W/mK
=density kg/m
3
=efficiency
r =radius m
SA =surface area m
2
t =time s
T =temperature
0
C
V =volume m
3
x =distance m

B) Subscripts
amb =ambient
des =desired
cool =cooling
f =fusion
i =inner
o =outer
pipe =cylindrical surface of pipe
tb =thermal battery
top =top surface of sealed pipe
scc =solar concentrating collector
sun (self-explanatory)

11
ACKNOWLEDGMENTS
The author wishes to thank all her family members for their unwavering support. In addition, she is
immensely grateful to Professor Wilson, for his guidance and mentorship throughout the duration
of the project. In addition, Jakob Hopping deserves many thanks for his assistance, insight, and
friendship.
Lastly, the MIT Electronic Research Society (MITERS) was particularly generous with supplies
and equipment, in support of this research project.

12
GLOSSARY
Absorber. A device that converts light energy to thermal energy.
Absorptance ratio. Ratio of absorbed solar radiation flux to the incident radiation flux.
AWS. American Welding Society.
Collector. Device used to capture solar radiation .
Concentration ratio. Ratio of the cross-sectional area of the incoming radiation to the cross-
sectional area of the radiation after concentration.
Concentrator. Device used to raise the temperature of incoming radiation, by focusing incoming
irradiation onto a reduced area.
Emissivity. Ratio of the emissive power of a surface to the emissive power of a black surface at
the same temperature.
Emitter. A body that radiates thermal energy.
Latent heat. Thermal energy stored or released as a material changes phase (solid/liquid or
liquid/gas) at constant temperature and pressure.
Latent heat capacity. Thermal energy stored or released per unit mass of material undergoing a
phase change.
ME. Mechanical engineering.
MIT. Massachusetts Institute of Technology.
MSDS. Material safety data sheet. It contains information on possible chemical hazards and safe
use of a material.
Phase-change material (PCM). A substance used for energy storage because of its change of
state, which occurs at a fixed temperature. Phase-change materials usually have large latent-heat
capacities.
Reflection. Change in direction of a wave front at an interface between two dissimilar media so
that the wave front returns into the medium from which it originated.
Sensible heat: Thermal energy stored or released as a material changes temperature.
Specific heat capacity. Thermal energy stored or released by a unit mass ot material per unit of
temperature.
Solar Cooker. Device that is powered by solar radiation. It is used to supply heat for preparing
meals.
Tracking. The mechanism whereby a device follows the apparent movement of the sun.
GLOSSARY

13
Thermal battery. A device that stores thermal energy to be released for use later.
Transmission. Propagation of light waves through a medium.
VITA. Volunteers in Technical Assistance.

14
Chapter 1) Introduction
Solar cookers convert solar radiation into heat, which is used to cook food. Such cookers - in
principle - require very low operating costs, because they rely on an energy source that is free,
abundant, and renewable. In addition, their operation inflicts virtually no damage on the
environment. Numerous designs (over 200) already exist. Many designs are described in books
1

and in the technical literature
2
. Even more are described on the internet
3
. A few are available
commercially and even fewer enjoy widespread use.
This low level of acceptance is primarily because most solar cookers have one drawback. Cooking
must occur when, and where, the sun is shining.
A) Mechanism of operation
Solar cookers rely on three physical processes:
i) Collection
The goal of collection is to direct solar rays towards an absorber surface. The process
begins when solar rays fall on a collector. The average solar iradiation is a physical constant
(1300 W/m
2
)
4
. Therefore, as shown in Equation 1, the gross amount of energy collected is
directly proportional to the surface area used for collection.
To maximize energy collection, rays must be perpendicular to the collector surface. This
requires tracking along two axes. The first axis follows the apparent daily movement of the
sun from east to west. The second axis follows the apparent yearly movement of the sun
from a horizon angle of L-23.45
o
to L+23.45
o
, where L is the local latitude. In Cambridge
(Massachusetts), the latitude is 42
0
, so the collector would oscillate yearly from 18.55
0
to
65.45
0
.
The technique used for transferring the collected rays to the absorber is determined by the
position of the absorber surface with respect to the collector. The two choices for that
transfer are reflection and transmission.
For high-temperature applications, concentration is necessary. Such temperatures are
facilitated by the reflection of solar rays from one surface onto a smaller area, so that the
intensity of the light energy is increased.
The three major types of concentrators are
2
:
(a) stationary and seasonally adjusted;
(b) line concentrators; and
(c) point concentrators.
Concentration ratios range from 200 for two-dimensional collectors to 40,000 for three-
dimensional collectors with tracking.

15
ii) Absorption
Once sunlight from the collector reaches the absorber, the light energy must be converted to
heat. The ratio of radiated light energy to absorbed heat energy is the absorption efficiency.
The two factors that govern this efficiency are the emittance ratio and the absorptance ratio.
The value of both factors increase with the temperature of the absorber. Overall efficiency
increases as the ratio of absorptance to emittance increases.
Two mechanisms can be used to increase the absorptance ratio.
(a) Fins
These trap solar rays by increasing the number of reflections each incident ray
undergoes, before finally being absorbed.
(b) Selective coatings
These utilize the same principle as fins, but on a microscopic scale. Owing to ease of
manufacture, selective coatings are used more widely than fins. Values of absorptance
ratio and emissivity for various materials are shown in Table 1
4
.
Table 1: Absorptance ratio and emissivity for selective coatings
NAME Absorptance ratio Emmissivity Ratio
Black coatings
Anodize Black 0.88 0.88 1
Carbon Black Paint NS7 0.96 0.88 1.09
3M Black Velvet Paint 0.97 0.91 1.07
Polyethylene Black Plastic 0.93 0.92 1.01
CONDUCTIVE PAINT
Brilliant Aluminum Paint 0.3 0.31 0.97
Epoxy Aluminum Paint 0.77 0.81 0.95
ANODIZED ALUMINUM SAMPLES
Black 0.65 0.82 0.79
Blue 0.53 0.82 0.65
Brown 0.73 0.86 0.85
METALS AND CONVERSION COATINGS
Black Chrome 0.96 0.62 1.55
Buffed Aluminum 0.16 0.03 5.33
Inconel X Foil (1 mil) 0.52 0.1 5.2
Stainless Steel
Polished 0.42 0.11 3.82
VAPOR DEPOSITED COATINGS
Aluminum 0.08 0.02 4
TAPES
4253M Aluminum Foil 0.2 0.03 6.67
Y93603M Aluminzed Mylar 0.19 0.03 6.33


16
As shown in Table 1, materials with dark colors (black chrome and black paint) have the
highest absorptance ratio. Such materials (with high absorptance ratios and high
absoptance:emmissivity ratios) are the most desirable for solar energy conversion
applications.
iii) Heat transfer and storage
Various mechanisms exist to store heat and transmit it to the food, so that heat losses are
minimized. The mechanism used is governed by the spatial and temporal position of the
food in relation to the absorber.
The three major classes of heat-storage technologies are listed below.
(a) Sensible-heat storage.
Heat is stored by raising the temperature of a material. The amount of heat stored =
mcT.
(b) Latent-heat storage.
Heat is stored by causing a change in phase - either from solid to liquid, or from liquid
to solid. The amount of heat stored = mH
f
.
(c) Thermochemical-reaction heat storage:
Most reactions either require energy absorption or involve the release of energy. Thus,
heat can be stored or released by reversing the direction of the reactions.

In many cases, two or more of the major physical processes may be combined in one entity. For
instance, in box cookers, the pot serves as the absorber and collector, eliminating the need for
radiation transfer from collector to the absorber.
Generally, cooking takes place at the same time and in the same location as absorption and
collection. By avoiding this restriction, the WSC offers virtually unprecedented convenience in the
realm of solar cooking.
B) Goals
Prior to the inception of this project, two students
5,12
working under the supervision of Professor
Wilson conducted analytical studies that led to recommendations for the construction of the WSC.
However, the concept was not tested since no prototype had been built. This research project was
embarked upon in order to validate the feasibility of the WSC, by building one prototype and
testing it. The following objectives were outlined.
i) Deliverables
On a warm, sunny, and dry day in Cambridge, Massachusetts - the WSC prototype would
collect sufficient heat so that even after four hours from the completion of solar collection,
it would still heat 1kg of water at room temperature to boiling point within 15 minutes.
Once the boiling point is attained, it would maintain that temperature for one hour.

17
The purpose of this project was to investigate of the practicability of the WSC. Therefore,
owing to time constraints some desirable features would have to be omitted if they did
not significantly impede basic functionality. If successful, thermal storage would be the
most significant advantage this cooker enjoys, in comparison with other solar cookers.
Thermal storage was therefore earmarked to receive the most attention.
ii) Planned features
(a) Total weight less than 80 pounds.
(b) Cost less than $50 US.
(c) Ease of manufacture in non-industrialized countries.
iii) Deferred features
(a) Salt-state indicator a mechanism that warns that charging is complete.
(b) Operator control of stovetop heat-transfer rate during cooking.
(c) Automatic tracking of the suns apparent movement.
C) Key Topics
Te following variables were of greatest importance in measuring the performance of the Wilson
solar cooker. Hence, experiments were designed to elicit the factors that control the following
parameters.
i) Heat-absorption rate during charging
The rate of net heat absorption by the cooker during thermal charging.
ii) Heat-loss rate during storage
The rate of heat loss during storage as a function of storage time and number of modules
stored.
iii) Heat-transfer rate during cooking
Rate and efficiency of thermal energy transfer from thermal battery to pot during cooking.
iv) Useful heat
The proportion of heat stored that can be transferred to a cooking pot.
v) Maximum system temperature
The highest temperature that the system can attain without negative effects.


18
First, the context for solar cookers is detailed in the background chapter. In the next chapter, the
methods used to design, build, and test the first WSC prototype are discussed. Subsequently the
observations made are presented in the results and discussion chapter. Lastly, those findings are
incorporated into recommendations for future design work.

Chapter 2) Background
A) History
Solar cookers have been in existence for centuries. The first patent is dated 1894
6
. Even though
there are more than two hundred functional designs, the cookers have not gained widespread
acceptance. This is surprising, because they offer many advantages. Firstly, they rely on a virtually
infinite energy source. Secondly, the user usually does not have to pay to utilize the energy source.
Thirdly, he does not have to travel far to gain access to sunlight. It streams into his residence on
normal days. Fourthly, the use of the devices does not degrade the environment.

Figure 1: Tracking box cooker
All of this is in contrast to the difficulty faced as part of the process of obtaining cooking
equipment and supplies, in industrializing countries. Firstly, low levels of income create challenges
in purchasing cookers. Secondly, traditional energy supplies are scarce, regardless of whether they
are in the form of petroleum by-products, coal, wood, or electricity. Lastly, the use of the above-
mentioned energy sources does have an unfavorable impact on the environment.

Figure 2: Fresnel solar cooker
Incidentally, places that fit this description, for example sub-Saharan Africa and Southeast Asia,
tend to enjoy high levels of solar insolation. Therefore, many of the listed problems could be
alleviated by solar cookers.

19
Why are solar cookers not being used more widely? Is it because they do not work? Are they
difficult to use? Do people dislike them? Might potential users be unaware of their existence?
Alternatively, are they too difficult to obtain? Perhaps they are too expensive.

Figure 3: VITA solar cooker
7
i) Temporal restrictions
From examining Figure 2 and Figure 3, it appears the answer leans most heavily toward the
third option, because the user is restricted to preparation of the food when the sun is
shining. Furthermore, the cooking must be done in an extremely unnatural fashion. Many
solar cookers call for suspending pots at the focal point of a reflector, or for placing the pots
in enclosures, as shown for the tracking box cooker in Figure 1. Monitoring of cooking and
addition of ingredients become more challenging with the pot inside an enclosure.

Figure 4: Parabolic solar cooker
Others require lengthy cooking times due to the relatively low operating temperatures. All
of these stipulations combined, place a cumbersome burden on the user, one that is rejected
very frequently.
Some parabolic solar cookers do attain relatively high temperatures. The specifications for
the parabolic solar cooker in Figure 4 predict a power output of 500 watts, which can bring
one liter of water to boil in less than thirty minutes.

20

Figure 5: Clear Dome solar cooker
The highest temperatures reported for solar-powered cookers is by Clear Dome solar, with
some products attaining 1371
0
C.
Since many families have hot meals for breakfast and dinner, the following question arises
from reviewing the five solar cookers mentioned above. How is food to be prepared in the
morning or at night, when there is no sun light? What about overcast days? Apparently,
users would need to have an alternative means of cooking. Those with low incomes the
target audience, do not easily meet this requirement for the possession of multiple cookers.
ii) Spatial restrictions
A secondary reason for the low level of acceptance is that solar cookers generally require
cooking outdoors. For many potential users, this is not easy to do. Consider the case of
apartment dwellers. Most share outdoor space with other apartment residents. In this and
many other situations, operation of a solar cooker would require sacrificing privacy.
Another problem with the requirement of cooking outdoors is inclement weather. To utilize
a conventional solar cooker on rainy or cold days, the user will likely be subjected to
physical discomfort.
12 years ago, Professor David Gordon Wilson (of the MIT mechanical engineering
department) designed a solar cooker that eliminates these requirements by converting
absorbed radiant energy to stored thermal energy for later use, possibly at a different
location. The Wilson solar cooker (WSC) can therefore be used at the convenience of the
user, under conditions of low or no insolation, for example a kitchen inside a building at
night.

21
Similar cookers, that store energy for later use, have been developed in the last four years.
In one study, acetanilide was used in a three-reflector solar cooker, with temperatures in the
range of 110 to 120
0
C
8
. The same authors also used erythritol with an evacuated tube solar
collector. The reported operating temperatures were similar
9
. In a third study, Murty and
Kaanthed
10
propose the novel concept of utilizing the phase-change material (transparent
lauric acid) as insulation. Reported temperatures were very low (42
0
C). Another
application was developed at the University of Salta, in Argentina. Solar heated aluminum
bars at 204
0
C, after transfer to an oven, were used to prepare food for up to forty people.

Figure 6: Stored-heat concentrating reflector in Salta, Argentina
iii) Temperature restrictions
The drawbacks to the last four systems combine to form the third reason for the low level of
acceptance of solar cookers. The drawbacks involve low temperatures, fixed temperatures
or both. A modern electric or gas cooker can boil one kg of water in five minutes. This is
certainly not the case with most solar cookers to date. To the contrary, they operate at
temperatures less than 120
0
C. Therefore, they typically require up to 180 minutes
7
to bring
0.15 kg of water to boiling and 420 minutes
8
to bring 10 kg of water to boiling. This slows
cooking time considerably - rendering it very unattractive to anyone who does not have the
time or willingness to wait two hours for a meal. This rules out many people.
In response, to increase cooking speed, the Wilson solar cooker is designed to operate at
temperatures up to 250
0
C with provisions for the user to adjust the temperature as desired.
B) Components
Consequently, Professor Wilson
11
proposed the following solar cooker design in order to address
the issues raised above. Those issues are addressed by designing the cooker to store absorbed solar
energy for later use.

22

Figure 7: Early Wilson solar cooker design
The proposed design specifies three primary components- a collector, a thermal battery, and a
stovetop.
i) Collector
The collector is made up of two reflectors. The primary reflector is a Fresnel mirror
constructed in accordance with instructions provided for the VITA solar cooker. The
secondary reflector is suspended at the focal point of the primary reflector. An aperture in
the Fresnel mirror permits transmission of the reflected rays to the absorber beneath.
ii) Thermal battery
The thermal battery consists of the insulated storage container and the storage modules.

23

24
(a) Storage modules
Each storage module consists of two metal plates welded together and filled with the
heat-storage material. The storage modules operate in two modes.
1. Absorber function. Charging phase.
Charging would normally occur during periods of high insolation. The top surface of the
highest module serves as the absorber, accepting radiation reflected from the secondary mirror.
To enhance conversion of solar radiation to heat, the absorption surface should therefore be:
colored black;
finned; and/or
coated with a highly absorptive material.
As this heat is conducted to adjoining modules, the temperature of the enclosed storage
material rises. If a phase-change material is used, the modules store sensible heat until the
melting point is reached, at which point energy is stored as latent heat. Once all of the salt is
melted, any further heat absorbed is again stored in the form of sensible heat as the molten salts
temperature rises beyond its melting point. The modules would be removed when fully charged,
to a separate enclosure optimized for storage and possibly, cooking.
2. Stovetop function. Discharging Phase.
When the collector and aperture cover are removed, a cooking pot can be placed above the
modules. At this point, cooking commences.
(b) Insulated storage container
The insulated storage container houses the modules. Its primary purpose is to prevent
heat loss from the modules during charging as well as during the time interval between
charging and cooking. The container consists of an inner and outer cylindrical casing,
separated from each other by an insulating material, such as foam. The opening at the
top of the container may be sealed, with an aperture cover, to reduce thermal losses. For
this, a transparent material (glass or plastic sheets) would be used, to permit reception
of solar radiation from the secondary reflector.
iii) Stovetop
When the collector is removed, a cooking pot can be placed directly atop the highest
module, at which point cooking commences.
C) Analysis
Two of Professor Wilsons previous students, Benjamin Matteo
5
and Chimba Mkandawire
12
,
analyzed the above design. As described in the next section, they derived pertinent operating
conditions and design values for the WSC.
i) Collection.
The factors that influence the energy collected are listed below.
(a) The concentration ratio
Higher concentration ratios reduce the absorber area required to absorb a given quantity
of radiant energy.
(b) Heat losses.
Four mechanisms exist for heat loss.
1. Reflection
Energy lost between incidence and reflection.
2. Transmission
Energy lost between transmission from the collector to the absorber.
3. Absorption
As shown in Table 1, most paints (like black chrome)
4
have an absorptance/emissivity ratio of
one. This requires a concentration ratio of 80 to achieve temperatures of 650K. In the process of
energy conversion from light to heat, some energy is lost.
4. Leakage
Energy lost from the system due to conduction and convection.

ii) Thermal charging
The time needed to collect a give amount of thermal energy is given by:
t
heat
=
scc sun scc
tb
A I
E

Equation 1: Module charging time

The following estimated values are used to calculate the time needed to collect a given
amount of thermal energy. The results are shown in Table 2.
(a) A solar collector efficiency of 50 %.
(b) An average solar irradiance of 1300 W/m
2
.
(c) A thermal energy requirement for each meal of 12.7 MJ .
This value is estimated by calculating the energy required to:
raise the equivalent of six liters of water from room temperature to boiling; and
maintain the boiling point temperature for one hour.

Table 2: Module charging time at 50% collection efficiency, with tracking
Collector Radius (m) Time (h)
1 0.84
0.75 1.49
0.5 3.37
Without E-W tracking, energy yield is up to 0.85% of that with full tracking.

25
iii) Thermal dissipation

Figure 8: Thermal conduction model

For the infinitesimal mass shown in Figure 8, the conductive heat-transfer rate is given by:

dx
dT
kA
dt
dE
=

Equation 2: Rate of thermal energy transfer

From equation 2, the time required for this energy to be dissipated is:
1
) (

=
dx
dT
kA
E
t
cool

Equation 3: Time required for thermal energy dissipation

The total thermal energy loss from the thermal battery was estimated using a pipe insulation
model, as depicted in Figure 9

Figure 9: Pipe insulation model

26
Integrating Equation 2 over the surface of the pipe gives
=
pipe
pipe
SA
E
i
o o
r
r
k
r
T
ln

Equation 4: Heat loss through a cylindrical surface

Summing up heat losses from the top, bottom, and vertical surface gives
the overall heat-loss rate = pipe
pipe
top
E
SA
SA
+ 2 1
Equation 5: Total rate of heat loss through pipe insulation

iv) Module geometry design
The energy stored by the module is related to the temperature by the following equation:
E
tb
=
LiNO
3
V
LiNO
3
(c
tb
T+H
f
)
Equation 6: Module energy storage

T is given by
T =T
desired
- T
ambient
Equation 7: Module temperature difference

v) Spring design
Dividing the total module weight by the solid height of the spring gives the required spring
constant.
Table 3: Spring geometry
Solid Height of Spring 1 in
Spring Constant 2850 N/m

D) Design
Upon completion of the analytical studies described above, Benjamin Matteo and Chimba
Mkandawire recommended the two designs shown in Figure 10 and Figure 11.

27

Figure 10: Recommended Wilson Solar Cooker: Version 1

Figure 11: Recommended Wilson Solar Cooker: Version 2

28

Chapter 3) Methods
Given the advances in materials development during the ten years since the above designs were
proposed, it seemed necessary to revisit some of those recommendations before construction
commenced. Figure 12 shows the WSC design proposed for this project.

Figure 12: Proposed Wilson solar cooker design
First, a material was selected to act as the heat-storage material. Next, a metal was selected for the
module container. Thirdly, the shape of the module container was designed. Fourthly, the module
assembly was developed. Simultaneously, the insulating container was designed and developed by
Jakob Hopping, an undergraduate student. Lastly, tests were formulated and executed.
A) Material selection
i) Heat-storage-material selection
(a) Heat-storage-material selection criteria.
The target users for the WSC were defined to be middle-aged women in rural areas of
aggressively industrializing countries, such as Malaysia. Ideally, these users would have
daily access to a space receiving solar insolation.
In addition, users would have someone to operate the device during absorption. This
could prove time-consuming (up to six hours per day).
Accordingly, a heat-storage material was sought that exhibited the following
characteristics, listed in order of precedence.

29
Chapter 3) Methods

30
1. Safety
There was a strong possibility that selected laboratory chemicals might be introduced into the
kitchens of people who are not necessarily familiar with such substances. Therefore,
mechanisms were sought to prevent those users from coming into bodily contact with, or
become unduly exposed to, chemical hazards.
In particular, the nitrate salts are known to pose specific safety risks. The nitrates of potassium
and sodium are commonly used as explosives. The material safety data sheet (MSDS
13
) for the
nitrates of potassium, sodium, and lithium specify the risk of explosions upon exposure to shock
or heat.
2. Low cost
Regardless of how well the product is designed, it serves no purpose if the users can not afford
it. Since the intended users are likely to earn relatively low incomes, and since the heat-storage
material would likely be the largest component of the system, low cost was sought for the heat-
storage material.
3. Readily available in developing countries
Importation of the heat-storage material would almost certainly increase the price of the WSC.
Furthermore, local availability of the heat-storage material would encourage local manufacture,
which would in turn enhance the local economy, which would then increase the buying power
of the target users. The goal is to help people in underprivileged countries by stimulating their
economy. Manufacturing the device in industrialized countries would only increase dependence
on foreign economies.
4. Low density
Minimization of the total weight of the WSC was a design goal. Since the latent-heat energy
stored is directly proportional to the mass of heat-storage material, it was important to find a
heat-storage material with the lowest density possible.
5. High latent-heat capacity
In order to reduce both weight and volume, materials with the highest specific-heat capacity
were sought.
6. High thermal conductivity
Once absorbed, thermal energy must be transferred quickly to the heat-storage material. Any
delays in absorption would increase dissipation, directly reducing efficiency of heat storage.
7. Melting point above 250
o
C
Operating temperatures above 250
0
C were needed for three reasons:
a to enable reasonable cooking speeds;
b to permit baking; and
c to permit frying.
Therefore, the melting point of the heat-storage material must also exceed that temperature
(250
0
C).
(b) Heat-storage-material candidates.
The following materials were considered.
1. Sensible-heat-storage media
These materials would fulfill the first three requirements. However, unlike phase-change
materials, the discharge temperature can not be kept constant.
Chapter 3) Methods
Table 4: Required volume and mass for selected sensible heat media
Material
Heat capacity (J /kg-K) Density (kg/m
3
) Required volume (cm
3
)
Required mass (kg) Material cost ($)
Ceramics:
alumina 1,045 3,980 7,637 30.40 28.88
clay 937 1,009 33,586 33.90
graphite 1,632 2,251 8,650 19.50
Cerium dioxide (CeO
2
) 171,069 7,280 26 0.20 3.76
Molybdenum disilicide (MoSi
2
) 41,840 6,250 122 0.80
Zirconium diboride (ZrBr
2
) 289,616 18 0.10 7.88
Non-ceramics
Concrete: 916 2,307 15,029 34.70

Furthermore, sensible-heat-storage media have low thermal conductivity. Therefore, their use
was removed from further consideration.
2. Phase-change materials such as the nitrate salts, tin, and proprietary compounds.
Eutectic mixtures of sodium nitrate and potassium nitrate were likely candidates because they
melt at desirable temperatures (such as 220
0
C) and had been used previously to store thermal
energy in solar applications.

Table 5: Thermal properties of nitrate phase-change materials
Material
Melting point
(
0
C)
Boiling point
(
0
C)
Density
(g/cm
3
)
Conductivity
(W/mK)
Specific heat capacity
(J /kgK)
Heat of fusion
(kJ /kg)
Potassium nitrate 333 400 2.1 0.5 267 118
Sodium nitrate 308 308 2.26 0.5 200 185
Lithium nitrate 250 2.38 385 367

Lithium nitrate emerged as the top candidate because it possessed a high melting point (between
250
0
C and 258
0
C) as well as a high latent-heat capacity - twice that of potassium and sodium
nitrate.

Table 6: Thermal properties of lithium nitrate
Melting point 623 K
Heat of fusion 367 kJ /kg
Specific-heat capacity 385 J /kgK
Conductivity 0.5 W/mK
Vapor pressure unknown

In spite of these favorable characteristics, lithium nitrate posed a safety risk since its material
safety and data sheet predicted that it would explode when exposed to heat, shock or strong
reducing agents. Measured values of the vapor pressure, as a function of temperature, are not
available.

31
Chapter 3) Methods
Table 7: Physical properties of solid lithium nitrate
Density
2.16 g/cm
3
Volume/kg. 0.46 l/kg
Weight required for the WSC to store 13 MJ 35.7 Kg / 78.6 lbs
Volume required for the WSC to store 13 MJ 15 l

In addition, the cost and availability of lithium nitrate were unfavorable. Lithium nitrate is sold
at the rate of $9/lb and is not available in most developing countries. It is available in South
Africa.

Table 8: Physical properties of liquid lithium nitrate
Density 1.78 g/cm
3
Specific volume 0.56 l/kg
Expansion ratio 21.4%

ii) Module-material selection
(a) Module-material selection criteria
In addition to the heat- storage material, a module-material had to be chosen. The
requirements were as follows.
1. Compatibility with the heat-storage material
Any reaction between the heat-storage material and module-material could lead to explosion or
degradation of either material. It could also lead to the production of undesirable products.
Repeated temperature cycling could not be permitted to cause degradation of either heat-storage
material or module-material.
2. Low density
Reduction of the module-materials density would lead to a proportional reduction in weight.
3. Low cost
Reduction in module cost was essential to satisfying the targeted total cost of $50 (US).
4. Available in developing countries
This enables local manufacture for intended users.
(b) Module-material candidates. The following materials were considered.
Table 9: Properties of module-material candidates
14
Steel Cast iron Mild steel Aluminum
Density (g/cm3)
7.9g/, 0.285 lbs/in
3
7.8 7.872 2.7
Coefficient of thermal expansion, linear 20C 16.6 m/m-C
Coefficient of thermal expansion, linear 500C 19.8 m/m-C
Specific heat capacity (J/kgK) 0.5 0.536 0.536 0.2241
Thermal conductivity (W/m-K) 16.3 80.4 50 167
Melting point (C) 1399 - 1421 C 660

32
Chapter 3) Methods
The three materials selected for testing were aluminum, mild steel, and stainless steel,
since they are widely available.
1. Aluminum
Aluminum had the advantage of lightness, low cost and ease of manufacture. According to the
MSDS, it posed the threat of reacting with lithium nitrate to create an explosion and form
undesirable products.
2. Stainless steel
Stainless steel held the promise of reduced reactivity and corrosion. However, its density of 7.9
g/cm
3
is almost thrice that of aluminum (2.7 g/cm
3
). Stainless steel also had the advantage of
ease of joining (through welding or brazing).
Two grades of stainless steel were considered.
a 304 stainless steel
This steel has high resistance to elevated temperatures, oxidation, and corrosion. However it
is difficult to machine. In view of the complexity of the proposed module design,
machinability was a significant factor.
b 303 stainless steel
This steel is more expensive, but is easier to machine. However, its weldability is lower. The
same factors that make this alloy corrosion resistant and strong also increase the difficulty
associated with welding it.
15
This led to discarding it as a module-material candidate.
Figure 13: Relative machinability of steels
16

iii) Compatibility research
(a) Theoretical predictions for compatibility between LiNO
3
and common metals
17

1. Pure lithium nitrate
The aluminum ion (Al
3+
) has a higher position in the activity series than the lithium ion (Li
+
).
Therefore, aluminum should not displace the lithium ion (Li
+
) in LiNO
3
. Based on this,
aluminum should not react with pure lithium nitrate.

33
Chapter 3) Methods

34
2. Commercial grade lithium nitrate
However, commercial grade lithium nitrate commonly includes such impurities as Na, K, Ca,
Al, Fe, CO
3
, Cl, SO
4
and H
2
O. These impurities accelerate the reactions between lithium
nitrate and container metals. Therefore, the likelihood of corrosion is directly dependent on the
concentration of these impurities. Thus, anhydrous LiNO
3
is reported as having high corrosivity
for thermal storage applications. Use of nitrates with aluminum was not recommended for
thermal storage.
3. Aqueous lithium nitrate
LiNO
3
inhibits corrosion of aluminum in aqueous alkaline solutions. This property is used in
recycled nuclear waste to reduce corrosion of aluminum. However, this effect is reduced at high
temperatures.
Also, mixtures containing LiNO
3
are used to anodize aluminum, which imparts resistance to
corrosion.
(b) Theoretical predictions for LiNO
3
stability.
Reported decomposition temperatures are 258
0
C (the melting point), 300
0
C, and 327
0
C. The
decomposition products are lithium nitrite and oxygen. Decomposition can be reversed in
oxygen at high temperatures.
Stability increases with mixtures of other nitrates. However, there are three disadvantages
associated with these eutectic mixtures. Firstly, they melt at lower temperatures. Secondly, their
heat of fusion is hard to find. Thirdly, nitrate eutectics eventually decompose with thermal
cycling.
After searching the literature it was unclear whether lithium nitrate reacted with either
stainless steel or aluminum, at temperatures below 350
0
C. Two researchers were identified
who had conducted experiments with nitrates.
The first was Dr. Robert Bradshaw
18
of the Sandia National Renewable Energy Laboratory.
He had used the nitrates of potassium and sodium with stainless steel, at temperatures above
300
0
C. He attested to the fact that no incidents had occurred and that he expected no
problems with the planned experiments.
Another researcher, Dr. David Kerridge, at the University of Southampton in the United
Kingdom
19
appeared to be the only one who had worked with lithium nitrate. His
predictions were much different. He said that experiments with nitrates had proved to be
very dangerous. In one case, a factory had exploded with many fatalities because an
incompatibility had been introduced unknowingly. Owing to the safety concerns raised,
aluminum was eliminated from further consideration.
iv) Compatibility tests
(a) Test 1: steel cup
This test was conducted to ascertain the compatibility of stainless steel with lithium
nitrate. It was also conducted to investigate the effects of thermal cycling and the by-
products of possible decomposition at high temperatures.
Chapter 3) Methods
To test stainless steel, a stainless steel cup was selected measuring three inches in
diameter and one inch in height. Its thickness was 0.050 in. To test mild steel, a mild
steel cap was selected with a thickness of 0.5 inch. The cup was filled with lithium
nitrate and sealed with the mild steel cap. The cap was tapered to ensure that no air,
liquid or solid escaped. For three weeks, the cup was heated and cooled from 280
0
C to
110
0
C. Each cycle was repeated every thirty minutes. This provided 504
heating/cooling cycle repetitions. A Corning laboratory heater PC-420 was used to
provide heat. A Grasslin digital timer switch was used for turning the heater on and off,
every thirty minutes.
(b) Test 2: flat pouch

Figure 14: Flat steel pouch test set-up
20

To address problems with leakage, brazing was identified as a joining method. In this
test, carbon steel was tested for use as the module container material. Two square 4130
mild steel sheets (nine inches wide, nine inches long and 0.025 inches thick) were
brazed together to form an envelope. This envelope was filled with one pound of
lithium nitrate and heated from room temperature to 250
0
C.
B) Module-shape design
The above-mentioned module-material tests had shown that there was no corrosion of either
stainless steel after 504 cycles (equivalent to eighteen months of use). Doubts about corrosion of
mild steel led to the initial choice of stainless steel as the module-material. Once material selection
was completed, the shape of the module was designed.
i) Module-shape criteria.
In view of the requirements for the completed cooker, the shape of the module had to meet
the following criteria.

35
Chapter 3) Methods
(a) High surface area to volume ratio
The rate of conduction of thermal energy, from the module container to the heat-storage
material, is directly proportional to the thermal conductivities of the module-body
material and the heat-storage material. The thermal conductivity of lithium nitrate is
low (0.5 W/mK). This is one-quarter the thermal conductivity of ice. This would lead to
poor conduction to the heat-storage material.
To account for this, the area of the module container surface in contact with the heat-
storage material was maximized by incorporating internal fins. Since conduction is
proportional to surface area, this would facilitate a higher conduction rate.
(b) Minimum weight and volume
To conserve the total volume and weight of each module, it was necessary to optimize
the amount of lithium nitrate enclosed in each module. In addition, there had to be
enough room left for the lithium nitrate expansion that would accompany the increase
in module temperature during charging.
(c) High absorption
The amount of heat stored by the module would be proportional to the amount of heat
absorbed. In turn, the amount of heat absorbed by the module would increase with the
area of the module used for absorption. Thus the module cross-sectional area used for
absorption needed to be optimized.
Another approach to maximizing heat absorption was to increase the absorptance ratio
by utilizing sprays, paints, gratings, and/or external fins.
(d) Portability
The module had to be designed in such a way that it was easy for the user to handle.
The initial version of the cooker might require that the user manually removes fully
heated modules and places them in a separate storage location. So each module had to
be designed so that the user would be able to gain access to it, and easily carry it for
more than 30 seconds at 300
0
C. This would require weights as low as four lbs and
handles which remained cool enough not to burn the user.
ii) Module-shape selection
Professor Wilson had originally proposed using two finned circular plates. The proposed
design is shown below.

Figure 15: Early module design

36
Chapter 3) Methods
This proposal was examined by two of his undergraduate students
5, 12
. They performed
thermal modeling and analysis, which resulted in minor modification to his proposal. Their
suggestions are shown below.

Figure 16: Revised module design

The original design and subsequent recommendations were used as the starting point for the
module design. The first modification was the removal of the external fins associated with
increased radiation absorption. In view of the complexity involved, incorporation of these
fins was deferred for a later stage of the project.
The design of the module was then carried out by following the two steps listed.
(a) Optimization of the surface area to volume ratio.
ProEngineer was used for the drawings and modeling. The surface area to volume ratio
for each design was computed. Subsequently, the module dimensions were revised in
order to increase that ratio.
(b) Allowance of enough room for lithium nitrate expansion upon heating and liquefaction.
The equation for lithium nitrate expansion
21
is listed below.
(T) =
m

T
(T-T
m
)
Equation 8: Lithium nitrate thermal expansion
The values for the constants in Equation 8 and the valid temperature range (T
min
, T
max
)
are listed below.
Table 10: Lithium nitrate thermal expansion constants
m
(g/cm
3
K)
T
(g/cm
3
K) T
m
(
0
C) T
min
(
0
C) T
max
(
0
C)
1.781 0.000546 253 441
1.93 0.000549 0 273 309

37
Chapter 3) Methods
Thus, the expected expansion is up to 21%, as the module is heated from room
temperature to 300
0
C.
Figure 17 and Figure 18 show three-dimensional and two-dimensional views, respectively
of the selected design.

Figure 17: Three-dimensional view of the design selected for the module top and bottom

The required module design parameters for the specified heating load of 13 MJ , are given
below.
Table 11: Module geometry design values (at room temperature)
Density of Steel 7.9 g/cm3
Density of solid Lithium nitrate 2.38 g/cm3
Single module
Diameter 15 in
Thickness (external height) 1 in
Weight of salt 4.1 kg
Volume of salt 1.9 l
Volume of steel 0.67 l
Volume of air space 0.23 l
Total volume 2.8 l
System
Number of modules required 8
Volume 23.2 l
Height 8 in

38
Chapter 3) Methods

Figure 18: Module-top design (stainless steel material, units in inches)

As shown in Figure 18, both the concave indentation of the module top and the internal fins
are preserved from the design proposed by Professor Wilson and his previous students. The
predicted weight of steel was five pounds. The predicted volume of lithium nitrate is 1.9 l
(4.1 kg).
C) Module construction
i) Material procurement & fabrication
(a) Module top
The selected design turned out to be too complex to fabricate at reasonable cost,
utilizing equipment at the MIT central machine shop or at the MIT Laboratory for
Manufacturing Productivity.

39
Chapter 3) Methods

Figure 19: Manufactured module top

Instead, Laurel Brooke, an external machine shop in New Hampshire, manufactured the
module tops. The top and bottom views of the product are shown above in Figure 19.
Type 304 stainless steel was ordered from Alliant Steel in New Hampshire. Aluminum
was provided by Laurel Brooke.
(b) Module bottom
Stainless steel module bottoms (shown in Figure 20) were purchased from Hamilton
Beach.

Figure 20: Module bottom

(c) Lithium nitrate
Lithium nitrate was initially purchased from Chemsavers at $50 per pound and then
from Chemetall at a cost of $8.75 per pound.
ii) Assembly
(a) Components
The complete module consisted of a module bottom, enclosed lithium nitrate and a
module top.
(b) Method choice
One challenge faced in assembling these components was maximizing the amount of
lithium nitrate enclosed. This was consistent with the goal of minimizing occupied
volume.

40
Chapter 3) Methods

41
Lithium nitrate is supplied in granular form. In this condition, it has the density of the
crystalline solid and 2/5th the density of the liquid. To minimize occupied volume while
leaving enough room for expansion upon liquefaction, the modules had to be filled with
lithium nitrate in the liquid state.
The granular form was not used because, for the same weight of lithium nitrate, the
volume occupied would be twice that of the liquid. The solid state was not used because
it would require very high pressures to pack the lithium nitrate between the modules
grooves.
(c) Method chosen
A sample of lithium nitrate weighing two pounds was placed into the module bottom.
Both module bottom and lithium nitrate were then heated to 300
0
C to melt and
compact the entire sample. While the lithium nitrate was still molten, the module top
(which was also heated) was placed over the bottom plate. The salt was allowed to
solidify as it cooled to room temperature. The top and bottom plates were then joined
and sealed.
iii) Joining and sealing
Once assembled the modules needed to be joined and sealed. The following methods were
considered.
(a) Welding
Welding facilities are widely available, even in developing countries. Moreover,
welding is relatively cheap and quick. Reasonable results can be produced by a novice.
One disadvantage of welding is the high temperature involved, since melting of the base
metal is required. Stainless steel, for instance has a melting point of 1400
0
C. The
material safety data sheet for lithium nitrate states that exposure of lithium nitrate to
high temperatures is an explosive hazard. However, this MSDS does not give a numeric
value for what it considers high temperature. It does provide a value of 600
0
C, as the
temperature beyond which decomposition occurs. The combination of decomposition
and/or explosion creates the possibility that high pressures will be developed in the
module. Welding of containers subjected to high pressures is not recommended.
Furthermore, welding creates high-temperature gradients in the region close to the
weld. Five minutes of heating at 600
0
C produces chromium diffusion in steel, which
will later lead to cracking. The chromium diffusion also reduces corrosion resistance.
(b) Soldering
Soldering involves the use of solder alloys which melt around 220
0
C. Since the desired
operating range for the module was as high as 300
0
C, soldering was not an option.
(c) Brazing
Brazing would offer lower temperatures than welding. As shown in Table 12, silver-
based brazing-alloys can be brazed at 760
0
C. This is still higher than the upper
temperature limit beyond which decomposition of lithium nitrate occurs. However, with
cooling in the region of the brazed metal, temperatures in the lithium nitrate could be
kept below 600
0
C during brazing.

Chapter 3) Methods

Table 12: Standard AWS brazing-alloy usage temperatures
9
AWS 5.8 Spec's F C
BAg-1 1145-1400 618-760
BAg-1a 1175-1400 635-760
BAg-2 1295-1550 702-843
BAg-2a 1310-1550 710-843
BAg-3 1270-1500 688-816
BAg-4 1435-1650 779-899
BAg-5 1370-1550 743-843
BAg-6 1425-1600 774-871
BAg-7 1205-1400 652-760
BAg-8 1435-1650 779-899
BAg-8a 1435-1650 779-899
BAg-13 1575-1775 857-968
BAg-13a 1600-1800 871-982
BAg-18 1325-1550 718-843
BAg-19 1610-1800 877-982
BAg-20 1410-1600 766-871
BAg-21 1475-1650 802-899

(d) Mechanical fastening
This would afford ease of assembly and disassembly. However, the addition of bolts,
nuts and a sealant (gasket or o-ring) would create an undesirable increase in the total
weight of each module.
Shield metal arc welding (SMAW), also known as stick welding, was used. It is a
commonly used process for containers with thick walls. The equipment is also easily
accessible. However, the process affords less control to the welder, resulting in lower
quality welds. Once briefed on the safety risks involved, and after being provided with the
MSDS for lithium nitrate, the welders at the MIT central machine shop and ME Pappalardo
laboratory were unwilling to perform the welding. To avoid exposing non-MIT welders to
the aforementioned risks, Professor Wilson performed the welding himself.
D) Assembled module tests
The total weight of the assembled steel module was 8 lbs. The heating experiments consisted of
heating the module from room temperature to a maximum temperature of 400 C. The module was
then removed and placed in an insulated storage container.
The following components were used.
i) Oven
Heat was supplied using a Thermolyne Furnatrol oven. The specifications list the maximum
temperature as 1200
0
C.

42
Chapter 3) Methods
ii) Insulated storage container
Built by J akob Hopping
22
, this container contained a metal plate for supporting the heated
modules. The plate was surrounded by fiberglass insulation. Elements of the design are
shown below in Figure 21, Figure 22 and Figure 23.
Figure 21: Insulated storage container

Figure 21 shows the manner in which the storage container is utilized during cooking. A pot
is placed on top of the storage modules, which can be raised or lowered by rotating the
handle.
Figure 22: Top and side views of the insulated storage container

43
Chapter 3) Methods
Figure 23: Metal drum used in construction of the insulated storage container:

iii) Temperature measurement
(a) Infrared thermometer
This thermometer utilized laser aiming, permitting temperatures to be monitored
wirelessly. The thermometer had the following specifications:
Built in socket accepted K-type or T- type thermocouples;
Measurement range from -60 to 500
0
C;
Probe ranges from -83 to 1400
0
C;
Ambient operating range from 0 to 50
0
C;
Accuracy 0.1
0
C; and
Response time 1 second.
(b) Multimeter:
This multimeter was used to measure raw thermocouple voltages. It also had a socket
for K-type thermocouples, which could be used to obtain temperature readings directly.
(c) K-type thermocouple.
Temperature was also measured using K-type wire thermocouples. Thermocouple
calibration was performed using the above-mentioned multimeter temperature readings.
(d) Data logging equipment.
1. Hardware
Signals from the thermocouples were amplified and conditioned using a 4-channel WinDAQ
module attached to a laptop via the serial port.
The WinDAQ module had the following specifications:
Four single-ended bipolar analog input channels;
10-bit precision;
+/- 10V ADC (analog to digital conversion) measurement range;
sampling rate up to 240 samples/second.

44
Chapter 3) Methods

45
Amplification was provided by an Analog Devices AD595 thermocouple amplifier. This chip
used an on-board ice point so there was no need to compensate for the ice point. The output was
10mV per
0
C.
2. Software
The laptop ran the WinDAQ software which displayed data in real-time and stored it for later
use. The laptop was used in the Windows operating system enviroment.
E) Module re-design.
i) Design cycle two
Based on observations from the design implemented above, the following changes were
made.
(a) Module material
The assembled module described above weighed eight pounds. That weight included
two pounds of lithium nitrate. This was inconsistent with design specifications of 80 lbs
for the maximum weight of the entire cooker. Two pounds of lithium nitrate in each
module would necessitate 8 modules (weighing 64 lbs) in order to cook a meal for a
family. To reduce this weight, aluminum was used in place of stainless steel for the
module body.
(b) Coating
As explained earlier, some sources predicted undesirable (corrosive and explosive)
reactions between aluminum and lithium nitrate. To prevent this, metal coatings were
considered, to provide a barrier between the interior surface of the module top and the
lithium nitrate. The module bottom was made of stainless steel, and so did not need to
be coated.
Various finishes considered include:
1. anodization;
2. electropolishing;
3. galvanization; and
4. thermal sprays.
Because of cost, thermal sprayed coatings are only used extensively in aerospace applications
They are used to a lesser extent, in commercial applications. However, they were chosen for
this application because of their versatility with respect to material and thickness. Almost any
material can be deposited using thermal sprays
23
. Stainless steel was sprayed on to the modules
by Falmer Thermal Spray in Lynn, Massachusetts. The total cost for materials and services
associated with the thermal spraying process was $200 per module.
(c) Shape design
Lastly, to improve manufacturability, the flange thickness Figure 18 was increased from
0.05 inches to 0.375 inches. Instead of one, five modules were manufactured. The same
(commercial, off-the-shelf) module bottom was used as in the first design cycle.
Chapter 3) Methods
(d) J oining
At temperatures above melting, the module tested above consistently leaked small
amounts of lithium nitrate. Because of this leakage, it was necessary to change, or at
least improve, the mechanism of joining the module top to the bottom. Instead of
brazing, the following joining methods were considered.
1. Mechanical fasteners
Mechanical fasteners would have the advantage of being easy to disassemble and reassemble. In
turn, this would improve the ease of manufacture and maintenance, especially since most target
users are in countries rapidly undergoing technological development. Thus, bolts were chosen
to fulfill this requirement.
The modules were bolted along the flanges, using 10 evenly spaced 3/8-bolts and matching
nuts.
2. Sealants
With the use of mechanical fasteners, sealants would be necessary for the complete prevention
of liquid, or gaseous, leakage of lithium nitrate. The following sealants were considered:
a O-rings
O-rings are commonly made of rubber. Thus, they would offer the most pliancy and
flexibility. However, little was known about the interaction between rubber and lithium
nitrate. The only guidance available was from the material and safety data sheet (MSDS),
which stated that lithium nitrate was incompatible with organic materials.
In addition, o-rings would require grooves on the module flanges. This would add
complexity to the manufacturing of the modules.
b Ceramic gaskets.
Ceramic gaskets would offer the advantage of simplicity, since grooves on the flanges were
not required. Another advantage stems from the fact that ceramics are relatively inert.
A ceramic (alumina) gasket tape was available from Richards Sensors at low cost. Its
specifications indicate that it was thermally stable up to 650
0
C. This gasket tape was tried in
one heating experiment. The material disintegrated upon interaction with lithium nitrate.
Consequently, ceramic gaskets were removed from further consideration.
c Metal gaskets
Of all sealant materials considered, the easiest outcome to predict was associated with the
use of metals as gaskets. This is because more was known about the interaction of lithium
nitrate with metals, than with other materials. In addition, previous tests conducted as part of
this study could shed light on the gasket metals being considered. The gasket shown below
was available from MDC vacuum products in Hayward, California. None of the dimensions
available matched the flange of the module being used for the test. Hence, the modules
would have to be redesigned to be able to take advantage of off-the-shelf gaskets.

Figure 24: MDC Copper gasket

46
Chapter 3) Methods

47
d Silicone sealants
Most silicone sealants were not suitable for use at the high temperatures (300
0
C) involved
in these tests. The highest rated material found in this category was Red Polyseamseal.
The manufacturer's specifications were as follows
24
:
1. permanently flexible from -85 to 500
0
C;
2. short-term exposure to 600
0
C ; and
3. seals and encapsulates heating elements in engines and most high-temperature
sealing applications.

After analyzing the four sealing methods described above, Red Polyseamseal (a silicone
sealant) was chosen because of the ease with which the modules could be assembled and
disassembled, as well as the ease of procurement.
(e) Assembly
The assembly was modified by preheating lithium nitrate and the module bottom in a kitchen
oven, instead of using a hot plate.
ii) Redesigned module test
Heating experiments were conducted using multiple modules, instead of one. The
experimental setup was the same as in Chapter 3, paragraph D.

48
Chapter 4) Results and discussion
Six modules were fabricated, assembled, and tested in conjunction with the insulated storage
container. This section presents the observations made during testing.
A) Health risks
In accordance with the MSDS, it was observed that lithium nitrate severely irritates the skin, eyes
and respiratory tract. Therefore, gloves and a mask were worn during testing in order to minimize
health risks.
B) Chemical and physical properties
The literature search had yielded unacceptably vague and sometimes conflicting predictions about
lithium nitrates compatibility with metals. Similarly, predictions about lithium nitrates stability at
high temperatures were inconclusive.
i) Test 1: stainless steel cup and mild steel cap
(a) Reactivity
1. Stainless steel
No corrosion was observed on the steel cup. This suggests that there was no significant reaction
between lithium nitrate and stainless steel.
2. Mild steel
Some oxidation was observed on the mild steel cap. This could be from reacting with the
lithium nitrate or from ambient air. In either case, the observed oxidation discouraged further
consideration of mild steel.
(b) Compatibility
Contrary to theoretical predictions, no explosion was observed, with the stainless steel,
mild steel or even with aluminum foil. All three metals were exposed to lithium nitrate
at temperatures as high as 300
0
C. Neither brazing nor welding of steel was
accompanied by explosions.
(c) Physical properties
Lithium nitrate appears to vaporize at temperatures as low as 300
0
C. As described in
the Methods chapter, the lithium nitrate was enclosed in a steel cup and covered with a
steel cap. Both were tapered, to form a liquid-tight seal. At the end of three weeks of
repeated heating and cooling (504 cycles), it was observed that 50% of the lithium
nitrate had escaped from the steel cup enclosure. This lithium nitrate was deposited on
the walls of the surrounding aluminum foil that was being used as a heat shield. Most of
the lithium nitrate deposits were found on areas of the foil above the steel enclosure. If
the escaping lithium nitrate was in the liquid form, it would simply have dropped to the
areas adjoining the steel enclosure. The only reasonable explanation for finding lithium
nitrate deposits at heights above the enclosure, is that the lithium nitrate vaporized and
condensed upon coming into contact with the aluminum foil barrier at lower
temperatures (perhaps during the cooling phase of the heating-cooling cycle).

49
This leads to the conclusion that lithium nitrate must have come into contact with the
aluminum foil at temperatures above its melting point. Since no evidence of corrosion
or explosions were observed, there seems to be a very strong possibility that aluminum
and lithium nitrate are compatible at temperatures below 300
0
C.
ii) Test 2: mild steel flat pouch
J akob Hopping conducted this test. He experienced difficulty heating the entire pouch to
melting. Possible explanations for this observation are detailed in the next section (thermal
energy storage results). Even though the material at the edges could not be melted, he was
able to melt the lithium nitrate at the center of the pouch. No corrosion or explosions were
observed.
This leads to the conclusion that mild steel is compatible with lithium nitrate. However,
mild steel oxidizes in air. Hence, unless it is coated with another material, its durability -
simply from exposure to atmospheric conditions - would be unsatisfactory for the desired
life cycle (more than a year) of the WSC.
C) Thermal property observations
As shown in Table 6, the critical thermal properties of solid lithium nitrate are listed below.
Heat of fusion 367 kJ /kg
Specific-heat capacity 385 J /kgK
Conductivity 0.5 W/mK
Equivalent tiquid state properties were not found in the literature.
Lithium nitrate turned out to be a very difficult material to work with during assembly. In both
tests two and three (above), it proved quite challenging to melt less than three pounds of the
material in a module using an 1100 W heater. The center areas receiving the most heat, melted
within 30 minutes. However, the areas beyond a diameter of four inches would not melt even after
an hour.
Three possible explanations arise.
i) High surface area to volume ratio
The pouch and the module were designed to have a very large surface area in comparison
with their total volume. The goal was to facilitate uniform distribution of heat within the
module via conduction.
The surface area to volume ratios of the module and pouch are 100 times greater that of a
cube. This has a significant impact on conducting heating tests with the module and pouch.
During charging, mechanisms have to be set up to prevent the high rate of heat loss which
would ordinarily result from the large surface area.
ii) Low conductivity of lithium nitrate
The thermal conductivity of lithium nitrate is low (0.5 W/mK). This is one-quarter the
thermal conductivity of ice.


50
According to the Equation 2, the rate of conduction of thermal energy from the center to the
edges is directly proportional to the thermal conductivity of lithium nitrate. Since low
thermal conductivity leads to poor conduction, it is possible for the material on the edges to
remain unmolten while the material in the center of the pouch/module melts.
iii) High latent-heat capacity
A third reason for the observed behavior is the relatively high latent heat of melting (367
kJ /kg) for solid lithium nitrate. The amount of heat required to melt a given mass of lithium
nitrate is proportional to its latent-heat capacity. The time taken to melt the lithium nitrate is
therefore directly proportional to the latent-heat capacity. Since lithium nitrate was chosen
for its high latent-heat capacity, it can be expected that the time required for melting would
also be high.

Expectations based on the last two factors (low thermal conductivity and high latent heat) are
contradicted by the observations that the molten lithium nitrate cooled very rapidly. There was
little time (minutes) between melting and re-solidification. The module top had to be in place
within that time interval. Otherwise the lithium nitrate would re-solidify, rendering it impossible to
wedge the fins in the lithium nitrate as the design called for (in order to minimize occupied
volume). In addition, if the emplaced module top was at room temperature, the lithium nitrate
would re-solidify upon contact with the module top (rendering it impossible to move the module
top to the correct mating position). Hence, the module top was pre-heated to the same temperature
as the melting point of lithium nitrate (258
0
C).
Such manipulation of the module components, at temperatures as high as 258
0
C, will definitely
pose an issue during manufacturing. There are significant risks involved that would invite further
re-examination of the assembly process.
D) Thermal energy storage results
i) Test 3: single steel module
Figure 25 shows the variation of temperature over time for a single module made up of
stainless steel and filled with two pounds of lithium nitrate.
The module was heated to 400
0
C and then placed in an insulated container. As shown on
the chart, the temperature dropped from 400
0
C to 200
0
C in three hours. The highest rate of
temperature drop occurred in the first 30 minutes. During that time, the freezing point was
reached. The rate of cooling slowed down as time progressed.

Figure 25: Temperature versus time in a stainless steel module
0
50
100
150
200
250
300
350
400
450
0 0.5 1 1.5 2 2.5 3 3.5
Time (h)
T
e
m
p
e
r
a
t
u
r
e

(
d
e
g

C
)

This is not the result expected. There should have been three distinct regimes, as shown in
Figure 26.
Figure 26: Theoretical predictions for temperature versus time in a pure substance cooled
at constant pressure

(a) Temperature drop above the melting point
Given that most sources record a melting point of 258
0
C for lithium nitrate, it would be
expected that the temperature would initially drop at a relatively constant rate, from
above melting to the melting point. The average slope during this initial temperature
drop would be expected to be negative.
Since the rate of heat loss is proportional to the temperature difference, this initial
temperature drop should be the fastest. This is in accordance with observations.

51
(b) Temperatures at the melting point
It would then be expected that the temperature remains approximately constant, long
enough to be detected on the temperature-time chart, until all the lithium nitrate
solidifies. No such slope change was observed in this test. The slope of the temperature-
time chart, within the interval five degrees above and below the melting point, is
approximately the same as the rest of the curve. This discrepancy raises the question of
how effective the modules would be for cooking at constant temperature or for storing
energy as latent heat.
(c) Temperature drop below the melting point
Once solidification is complete, the temperature drop would be expected to continue
until room temperature is reached. The observed values for the temperature drop after
melting, are shown below in Figure 27. As expected from Equation 4, approaching
room temperature causes the rate of temperature drop to decrease exponentially.
Figure 27: Temperature versus time of lithium nitrate in a stainless steel module (after
solidification)
0
50
100
150
200
250
0 5 10 15 20 25 30 35 40 45 50
Time (h)
T
e
m
p
e
r
a
t
u
r
e

(
d
e
g

C
)

Figure 28 shows results from a similar test, but with a different kind of insulation.
Instead of fiberglass, sawdust was used. The temperature drop from 200
0
C to 100
0
C is
much faster than is shown in Figure 25. This can be attributed to the increase in thermal
conductivity of the loosely packed sawdust, as compared with fiberglass.

52
Figure 28 also provides information on the rate of heat conduction within the module.
Temperature versus time measurements were collected from two locations on the
module. The first location was dead center on the bottom of the module. The second
location was on the top edge. As shown on the chart, throughout the period of data
collection (three hours), a temperature difference of 15
0
C was observed between the
two thermocouple locations.

Figure 28: Temperature difference across a single steel module in sawdust insulation
0
50
100
150
200
250
0 0.5 1 1.5 2 2.5 3 3.5
Time (s)
T
e
m
p
e
r
a
t
u
r
e

(
o
C
)
Edge
Center

This is in conflict with the known value of thermal conductivity for stainless steel,
which is 16.3 W/m-K. A temperature difference of 15
0
C across a module 10 inches in
diameter and one inch thick would not be expected.
ii) Test 4: two aluminum modules
In this test, one module was heated to 300
0
C and placed in the insulated storage container.
Three hours later, another module heated to 300
0
C was added to the insulated storage
container. Subsequently, the temperatures of both modules were recorded.
Figure 29 shows that the new modules temperature started at 300
0
C, dropped to 240
0
C
and remained constant for approximately six hours. Compared with test three (single
module), this behavior is more consistent with expectations for temperature variation during
storage.
The initial rate of temperature drop is the highest observed throughout the test. This is
in accordance with Equation 2, which predicts that heat loss would be proportional to
the temperature difference between the module and its surroundings.

53
Figure 29: Temperature versus time for two modules in a pre-heated storage container
0
50
100
150
200
250
300
350
0 10 20 30 40 50
time (h)
T
e
m
p
e
r
a
t
u
r
e

(
d
e
g

C
)
60

It was observed that the temperature remained approximately constant long enough to
be detected on the temperature-time chart (in this case four hours), until the entire
lithium nitrate solidified. This is in accordance with predictions for the usefulness of the
modules during cooking. This chart confirms the feasibility of cooking food at constant
temperature of 240
0
C for four hours.
At that point, the temperature drop would be expected to continue until room
temperature is reached. The observed values for temperature drop to room temperature
are shown above in Figure 29. As expected, approaching room temperature causes the
rate of temperature drop to decrease exponentially. The temperature stayed above 125
0
C for 25 hours. It remained above 50
0
C for more than 50 hours.
iii) Test 5: three aluminum modules
In this test, two modules heated to 300
0
C were placed in the insulated storage container.
Three hours later, another module heated to 300
0
C was added to the insulated storage
container. Subsequently, both the temperatures of the initial modules and that of the new
module were recorded.
Figure 30 shows that the last modules temperature started at 270
0
C and remained constant
at 250
0
C. It did not drop to 240
0
C until after six hours. Compared with test four (two
modules), this behavior is even more consistent with expectations for temperature variation
during storage.

54
Figure 30: Temperature versus time for three modules in a pre-heated storage container

0.00E+00
5.00E+01
1.00E+02
1.50E+02
2.00E+02
2.50E+02
3.00E+02
0 5 10 15 20 25 30 35
Time (h)
T
e
m
p
e
r
a
t
u
r
e

(
C
)
Initial module
New module

As in the two previous tests, the initial rate of temperature drop before solidification
was higher than after solidification. This is consistent with expectations.
In comparison with test four (two modules), the period of solidification is longer six
hours. This is to be expected since more modules (three) were used in this test.
As expected, approaching room temperature caused the rate of temperature drop to
decrease exponentially. The temperature remained above 125
0
C for 30 hours.
E) Key system parameters
Utilizing the observations recorded in the preceding paragraphs of this section, it is now possible to
evaluate some of the system parameters identified as key, at the beginning of this project.
i) Heat-absorption rate during charging
The concentrator and absorber were not incorporated into this project, owing to time
constraints.

55
ii) Heat-loss rate during storage
The rate of heat loss depends on three factors.
(a) Heat-storage phase
Heat loss was calculated by using the rate of temperature drop in the modules, as well
as the mass of lithium nitrate in the modules. The mass of the module container was
ignored.
Table 13: Heat-loss rate as a function of heat-storage phase
mass of lithium nitrate (kg) 1.4
latent heat phase
time required for solidification (h)latent heat of fusion (J /kg) heat loss rate (W)
6 367000 23.79
sensible heat phase
temperature drop rate (
0
C/h) specific heat capacity(J /kgK) heat loss rate (W)
5 385 0.75

As shown in Table 13, the rate of heat-energy loss during melting was 30 times that
observed upon re-solidification. This ratio is far higher than would be expected, since
the insulation is the same. The only possible reasons for the difference (between latent-
and sensible- heat-storage loss) are listed below.
1. Higher thermal conductivity
The liquid lithium nitrate in the molten state had a higher thermal conductivity than in the solid
state.
2. Heat losses through convection
In the liquid state, lithium nitrate could lose heat through convection as the salt moves inside the
module.
Table 14: Predicted heat-loss rate as a function of temperature difference.
Temperature difference (K) Heat loss rate (W)
20 8.81
40 13.44
60 18.11
80 22.75
100 27.42
120 32.08
140 36.72
160 41.39
180 46.03
200 50.69
220 55.36
240 60.00
260 64.67
280 69.33
300 73.97
320 73.97
340 83.28
360 87.94

56
3. Higher temperature differences
In the molten state, temperatures were higher than in the solid state. The resulting temperature
differences (between the module and its surroundings) led to higher rates of thermal conduction
and thermal energy loss in the molten state. This agrees with predictions, as shown in Table 14,
that the heat-loss rate increases at higher temperatures.
Neither factor accounts for the large discrepancy between the latent-energy heat-loss
rate on one hand, and the sensible-energy heat-loss rate on the other.
(b) Number of modules stored
For the module test results shown in Table 15, additional modules were placed in the
storage container at different times, approximately three hours apart.
Table 15: Heat-loss rate for varying numbers of modules
mass of lithium nitrate per module (kg) 1.4
latent heat of fusion (J /kg) 367000
Number of modules time required for solidification (h)total heat loss rate (W)
1 0.5 285.44
2 4 35.68
3 6 23.79

This was done to simulate the manner in which the modules would be removed from
charging, at intervals of approximately one hour.
The number of modules stored is a very important variable when it comes to the rate of
heat loss from the system. Table 15 shows that the first module lost its thermal energy
ten times faster than the third module.
This is to be expected, since the rate of heat lost through thermal conduction is directly
proportional to the temperature difference (between the module and its surroundings).
The temperature difference was lowest in the case of three modules.
(c) Insulating material
Table 16: Predicted heat loss as a function of insulation thickness.
Insulation thickness (in) Heat loss (W)
7 87.
8 84.
9 82.
10 80.97
11 79.86
12 79.14
13 78.67
14 78.31
15 33.33
16 78.36
17 78.50
18 78.75
19 79.08
20 79.47
21 79.92
22 80.42
23 80.94
24 81.53
75
69
50

57
Table 16 shows the results of predictions made by a previous student for the heat-loss
rate as the insulation thickness is varied. Table 17 shows that the nature of the
insulating material has a significant effect on the heat-loss rate. In this test, the
insulating capacity of the sawdust was significantly reduced because it was loosely
packed.

Table 17: Heat-loss rate for different insulating materials
Mass of lithiumnitrate (kg) 1.4
Specific heat capacity(J /kgK) 385
Insulation material Thermal conductivity (W/mK) Temperature drop rate (deg/h) heat loss rate (W)
Fiberglass 0.04 12.5 1.87
Sawdust 0.042 25 3.74

iii) Heat-transfer rate during cooking
No tests were conducted to determine the rate and efficiency of thermal energy transfer
from thermal battery to pot during cooking. This would be the next step recommended for
future tests.
iv) Useful heat
Even though the results for temperature versus time were sometimes surprising,
observations on the proportion of heat stored that can be transferred to a cooking pot are
not. Table 15 shows that heat is lost from each module at a rate of 23W, during
solidification. If averaged throughout cooling, the fraction of useful heat lost is 3% per
hour. Figure 30 shows that the temperature remains relatively static at above 258
0
C for
more than six hours. Furthermore, once the temperature drops below the phase-change
point, it remains higher than 100
0
C for more than 25 hours. It is therefore to be expected
that the WSC would be able to supply the 13MJ needed to cook a three-pound meal many
hours (at least six) after charging is complete.
v) Maximum system temperature
As shown in the above graphs, if the system is leak-proof, it can attain temperatures as high
as 400
0
C without any negative effects. The Red Polyseamseal sealant was the only
component degraded during operation at this temperature.
This provides some answers to the questions raised by the vague and sometimes conflicting
data in the literature about thermal decomposition and stability of lithium nitrate at high
temperature. It suggests that 400
0
C is not high enough to cause any of the predicted
explosive hazards.

Since measured values of the vapor pressure as a function of temperature are not available,
tests should be conducted to ascertain the vapor pressure developed during heating.

58

59
Chapter 5) Conclusion and Recommendations
After completing the analysis, design and testing of the heat-storage system, the following
suggestions are offered to assist with future work on this project.
A) Design
i) Materials selection
(a) Heat-storage-material
Lithium nitrate, the heat-storage material, has been shown to meet the stated
requirements of storing heat at a constant temperature of 258
0
C for up to six hours.
Furthermore, this heat-storage material stores heat at temperatures above the boiling
point of water, for up to 25 hours. Thus, it is expected that a meal for up to six people
can still be prepared six hours after charging.
(b) Module-body material
1. Weight
The choice of material for the module body has a significant impact on the weight of the
module and of the entire system. The assembled steel modules weighed eight pounds. The
assembled aluminum module weighed four pounds. If users are to be expected to manipulate
these modules between the charging and cooking phases, the steel modules are too heavy. The
weight of the aluminum modules is more acceptable.
2. Availability
Both aluminum and steel are widely available in virtually every country. To reduce costs,
recycled metals could be used
3. Compatibility
The observation that lithium nitrate may be compatible with aluminum at temperatures below
300
0
C deserves further examination. Since the density of aluminum is one-third that of steel,
this would lead to a proportional reduction of the total weight of the WSC. This would also
eliminate the need for thermal spraying, which costs more than $200 per module. Even at
commercial production rates, because of cost, spraying would not have been a feasible option.
In view of the above observations, the recommended module-material is aluminum. It
has the lowest cost and the lowest density. It also is the easiest to manufacture.
ii) Module construction
(a) Fabrication
The current weight (four pounds) of the aluminum modules is acceptable. However, it
can be reduced by using (rolled or drawn) aluminum sheets, instead of machined
aluminum. This would possibly cut the weight, and cost, by up to one-third of the
current values.
The current module shape (with internal fins) requires advanced manufacturing
equipment readily available, only in industrialized countries. Once a prototype is
available, casting can be used for production runs.
Conclusion and Recommendations

60
(b) Assembly
The current assembly process requires exposure of the assembler to very high
temperatures of 258
0
C. This might be addressed by automating the process, if demand
rises to volumes high enough to sustain the additional cost.
(c) J oining and sealing
Mechanical fasteners (in conjunction with sealants) are easy to assemble and
disassemble. However, welding and brazing have the advantage of being much more
accessible in most countries of interest.
iii) Module-shape design
As mentioned earlier, the internal fins would be difficult to manufacture in non-
industrialized countries. A simpler shape, such as a thin flat disk, would significantly widen
availability of manufacturing equipment.
The current design calls for users handling modules at high temperatures (258
0
C).
Professor Wilson
25
has proposed adding handles made of a material with very low thermal
conductivity (such as foam or ceramic glass). These handles would maintain temperatures
that a user could touch without being harmed.
The temperatures in question involve a high level of risk as far as user safety is concerned.
From discussions with Professor Wilson
24
, it appears that it will be worthwhileto eliminate
the possibility of users being exposed to the modules at such high temperatures. The system
could be designed so that the absorbed thermal energy is conducted through metal pipes or
rods to the insulated modules during charging, and released from the modules in the same
position during cooking. The user would not have to move the modules in between charging
and cooking, when the temperatures are highest.
B) Performance
Tested components of the Wilson solar cooker display the potential for performing satisfactorily,
storing heat at a constant temperature of 258
0
C for up to six hours.
The quality of insulation alters the heat-storage-capacity period of the system by more than one
order of magnitude. For instance, changing from fiberglass to sawdust cuts the storage time by
half.
Referring to Table 18, fiberglass provides adequate insulation. However, it poses health risks,
because it inflames the respiratory tract. Styrofoam melts at low temperatures. Aerogel may not be
available in non-industrializing countries. Concrete or brick are commonly available and are
recommended for further use as the insulating material.
Continuously maintaining the storage container at temperatures above room temperature would
also greatly improve the heat-storage capacity of the system
The next set of tests should be conducted with heating loads (food to be cooked) added to the
system.


61
Table 18: Thermal conductivity of common materials
Material Thermal conductivity (W/m K)*
Silver 406
Aluminum 205
Glass,ordinary 0.8
Concrete 0.8
Water at 20 C 0.56
Asbestos 0.16
Brick,insulating 0.15
Wood 0.13
Snow (dry) 0.12
Saw dust 0.042
Fiberglass 0.04
Cork board 0.04
Wool felt 0.04
Rock wool 0.04
Air at 0 C 0.024
Aerogel 0.016
Styrofoam 0.01

C) Deferred features
Some of the features deferred at the beginning of the project are listed below.
i) Salt-state indicator
ii) Operator control of stovetop heat-transfer rate during cooking.
iii) Automatic tracking of the suns apparent movement.

The following components were deferred for later research as the project progressed.
iv) Radiant energy collection
v) Thermal energy absorption
vi) Cooking


62
As mentioned in the background section, commercial parabolic solar cookers do attain relatively
high temperatures. In the mid-temperature range, a power output of 500 watts can easily be
attained. Since the Clear Dome solar cookers temperatures are as high as 1371
0
C, attaining a
desirable heat-absorption rate becomes merely a question of finding an appropriate absorber and
developing a tracking scheme. In comparison with development of the heat-storage mechanism, it
is to be expected that development of the concentration and absorption scheme will be much more
straightforward.

63

REFERENCES

1
Bowman, T. Understanding Solar Cookers. Arlington: Volunteers in Technical Assistance; 1985.
2
Wyman, C. A review of collector and energy storage technology for intermediate temperature
applications. Solar Energy. 1980; 24: 517-540.
3
Solar Cookers International. The Solar Cooking Archive. Available at http://www.solarcooking.org.
Accessed 9 August 2004.
4
K&K Associates. Thermal finishes. 2005. Available at http://www.tak2000.com/data/finish.htm.
Accessed 27 May 2006.
5
Matteo, BC. Stored-heat solar cooker. (SB Thesis, Massachusetts Institute of Technology;
1995).
6
Radabaugh, JM. Heavens flame Ashland: Home Power; 1998.
7
Solar Cooker Construction Manual. Arlington: Volunteers in Technical Assistance; 1967.
8
Buddhi D, Sharma SD, Sharma A. Thermal performance evaluation of a latent heat storage unit
for late evening cooking in a solar cooker having three reflectors. Energy Conversion and
Management. 2003; 44: 809-817.
9
Sharma SD, Iwata T, Kitano H. Experimental results of evacuated tube solar collector for use in
solar cooking with latent heat storage. Proceedings of the EM4 Indoor Workshop IEA ECES IA
Annex, 2004; Indore, India. Available at
http://www.fskab.com/Annex17/Workshops/EM4%2520Indore%25202003-03-21--
24/Presentations/SDSharma%2520PAPER.pdf. Accessed 30 April 2006.
10
Murty, VV. Thermal performance of a solar cooker having phase-change material as transparent
insulation. Proceedings of the EM4 Indoor Workshop IEA ECES IA Annex, 2004; Indore, India.
Available at http://www.fskab.com/Annex17/Workshops/EM4%20Indore%202003-03-21--
24/Presentations/VVS_murty.pdf. Accessed 30 April 2005.
11
Wilson, DG. The Wilson Thermal-Storage Solar Cooker. (Concept Memo and Pre-proposal,
Massachusetts Institute of Technology; 1997)
12
Mkandawire, C. Designed and Modeled Solar Cooker. (SB Thesis, Massachusetts Institute of
Technology; 1994).
13
Mallinkrodt Chemicals. Material safety data sheet L6958. Available at
http://www.jtbaker.com/msds/englishhtml/l6958.htm. Accessed May 30, 2006.
14
Automation creations. Material property data. Available at http://www.matweb.com. Accessed April
25, 2006.
15
Palmer, J. Brazing and welding 304L stainless steel. Brewing techniques [serial online] 1994;
2:6. Available at: http://www.brewingtechniques.com/library/backissues/issue2.6/palmer.html. Accessed
April 25, 2006.
16
Stainless steel- fabrication. Available at http://www.azom.com/details.asp?ArticleID=1178. Accessed 27
Apr 2006.
17
Lovering, DG (Ed). Molten salt technology. New York: Plenum Press; 1992
18
Phone conversation, August 2004 with Dr. Robert Bradshaw
19
Phone conversation, January 2005 with Dr. David H. Kerridge
20
Hopping, J A. Solar Cooker Progress/Data. (UROP rogress report, Massachusetts Institute of
Technology; 2005).
REFERENCES

64

21
Washburn, EW. Knovel International Critical Tables of Numerical Data, Physics, Chemistry
and Technology (1st Electronic Edition). 1926 - 1930; 2003: pp. 24). Online version available
at:http://www.knovel.com/knovel2/Toc.jsp?BookID=735&VerticalID=0. Accessed April 25, 2006
22
Hopping, J A. Development of a solar cooker (UROP Report, Massachusetts Institute of
Technology; 2005).
23
Thermal Sprayed Industrial Coating. Engineers Edge. Available at
http://www.engineersedge.com/finishing/thermal_spray_coating.htm. Accessed May 13, 2006
24
Barr, TF. Red high-temp silicone sealant. Mentor: OSI Sealants; 1999
25
Email correspondence with Professor Wilson, May 15, 2006

Development of An Asynchronous Solar-Powered Cooker

Diunggah oleh

Informasi Dokumen

Judul Asli

Hak Cipta

Format Tersedia

Bagikan dokumen Ini

Bagikan atau Tanam Dokumen

Opsi Berbagi

Apakah menurut Anda dokumen ini bermanfaat?

Apakah konten ini tidak pantas?

Hak Cipta:

Format Tersedia

Development of An Asynchronous Solar-Powered Cooker

Diunggah oleh

Hak Cipta:

Format Tersedia

1

Anda mungkin juga menyukai